[MV] How do you Train Dictate when...

[Chuck Rogers]

You need to edit the choice in the recognition window that is closest to what you actually said. To do this, say "edit" followed by the number that is to the left of the choice that is closest to what you actually said. This will allow you to edit the text left of the number. Change the text to what you actually said using your keyboard and mouse (you could re-speak it, but it got it wrong the first time, so it probably won't get it right if you say it again). If you are unsure of what you actually said, click the "Play" button in the upper left corner of the recognition window. When you are finished editing, say "pick" or "choose" followed by the number of the choice you just edited.

So if choice #3 is closest to what you actually said, here is what you would do:

1). say "Edit 3"

2). Manually fix the text next to #3 to match what you actually said.

3). say "Pick 3" or say "Choose 3."

Regarding your global text macro, my guess is that you are not pausing before and after saying the name of the command. Dictate needs a little time both before and after a spoken command to realize what you are saying is something upon which it should act rather than something it should type. If pausing before and after doesn't fix the problem, check to make sure what you are saying is EXACTLY what is in the "Command Name" field in the Command Editor. (In this regard, it is not Dictate that needs to be improved, but the way you use it.)

If you are interested, The Emergent Institute offers one-on-one training on MacSpeech Dictate. They can be contacted through their web site (www.theemergentinstitute.org).



On Sep 10, 2010, at 10:24 PM, DealTek wrote:

> How do you train Dictate -- when you say a sentence, and the recognition window lists multiple choices of possible correct versions, 
> however none of them are correct?
> 
> 
> Also, along the same line ... 
> 
> I have created some local user defined text entries and they also do not appear correctly in the recognition window like:
> 
> global text command : say "print my name" but print "Mr. Joe G. (CoolGuy) Smith III"
> 
> As an example I might say: "print my name" - and the global text should print "Mr. Joe G. (CoolGuy) Smith III" - but instead it prints the actual words: "print my name"
> 
> How do you correct and improve dictate in this regard?